Version 3 Tenuto

• Jan 20, 2019 - 15:10

Updated to Version 3 this morning and the first issue is that I can't get a tenuto articulation time stretch to "stick". It always immediately reverts back to the default .10. Is this a bug or has the procedure changed to make these work?
Thank you


Comments

Now I find that all fermata time stretch settings on ALL my scores has be set back to the default of 1.00. This is a disaster! How can I get these settings back to where they were before the update?

In reply to by Jojo-Schmitz

OK, I'm going back to work using Version 2.3.2 (with the buggy tenutos); however, I made the mistake of saving the score I'm working on in Version 3, and now when I try to open it in Version 2.3.2 I get the error message:
"XML read error at line 243 column 18 KeySig". Is there any way to locate and fix that?

In reply to by hmscomp

It's a bug if it works different than intended, and AFAIK it was never intended that accidentally changing the stretch property for markings other than fermatas would have unintended consequences on playback. Sounds like you discovered a way of exploiting this bug by doing this deliberately, and I agree it seems convenient, so feel free to file a Suggest to the issue tracker requesting it be supported officially as an actual feature.

Meanwhile, the supported way of altering note lengths remains the Piano Roll Editor, which has been greatly improved in 3.0

So, I'd like to clarify: were you using the time stretch on tenutos to get them to do their normal function of making notes more legato, or were you really wanting them to act as fermatas? Because that's what time stretch was really doing - changing the tempo at that spot in the same way as a fermata. Which is why it was originally considered a bug - it was assumed that people would not want to accidentally introduce pauses into their piece when they were really just trying to make things more legato. It never did the legato thing, and it would be natural to want it to, and indeed, I'm totally in favor of adding a new property to articulations to do what most people were probably assuming time stretch would do anyhow.

But it has now come to our attention that some people really do want tenutos to act exactly like fermatas - to introduce small pauses into the playback, not just to make the notes more legato. And now I'm trying to get a sense of how prevalent this practice is. The workaround is to use invisible fermatas, and while it does work, it is admittedly a bit painful if you do this a lot. But I'm trying to weigh that against the pain of what may end up being a relatively large and risky change to the playback and/or layout code to reintroduce this property for articulations. Formerly, fermatas were articulations and thus in the same class (literally, in the C++ sense) as tenuto, but now they are separate classes, and handled very differently at different points in the code.

In reply to by Marc Sabatella

I want (more accurately, "have wanted since time began") an inspector property to say how much of a note, to what per mille (1/1000) each articulation cuts back, eg, 500, 850, 960, even 1100, whatever. To make it into "virtual fermata" would be to preclude its ever being done "right". The playback engine already respects the hard-coded percent/permille of articulations; I have made editing the .mscz to customize it into an art, and it never fails. More appropriate UI than these text-edits is needed.

Perhaps time-stretch and cutback should just be two different properties.

In reply to by [DELETED] 1831606

Internally, they are two different properties. I don't see any obvious reason we couldn't expose the "gate time" property (which controls what you call the "cut back" - it's basically a percentage of nominal duration) with very little code change. And this allows the "normal" case of staccato and tenuto to be tweaked. Similarly for "velocity" to be used for accents. But the "time stretch" property - originally intended for fermatas only - is no longer there to be exposed, which is why its greyed out. Even though it's a relatively uncommon usage, a few people apparently do rely on it. So adding that back is worth considering. To me, a lower priority than just directly supporting the "normal" properties for articulations, but both could conceivably be done at once.

Currently, the normal properties (gate time and velocity) are set in instruments.xml both globally and per-instrument, no way to override them per-articulation.

In reply to by Marc Sabatella

Is it contemplated, in a future version, the real playing effect to legato???

I'm not only talking about the time duration of the notes.

In the real world, some wind instruments (and the human voice, too) can change from a note (let's say C4) to the next note (D4) without any time pause and adding a little "glissando" effect (which is absolutely possible from any MIDI device).

Is it contemplated this feature???

In reply to by jotape1960

The "slur" mark that indicates legato "actually works" with some instruments, such as flute and piano, both of which do not play to 100% duration unless it is used. I have not figured out how, other then looking at instruments.xml, to tell which instruments do this, or how to turn it off (is it documented?). Other instruments still sound detaché even at 100% of duration. In this score, https://musescore.com/bsg/scores/5159652 I have "hacked" tenuto to mean 110%, put it on all notes and hidden it, and removed it on selected notes. There should be better ways, and the lack of consistency of legato handling is a trap.

In reply to by Marc Sabatella

I always thought a tenuto was defined as to hold the note/chord for just a little longer (but less than a fermata) than the stated tempo for expressive purposes. So yes, I was using them to "time stretch". I never heard of them as having anything to do with legato. But I'm no expert.

In reply to by [DELETED] 1831606

Indeed, this is why it is so surprising to me that this is a thing at all - I've never heard of it, and I shudder to think of the ramifications for ensembles when used in this way.

But in theory, if we have separate time stretch and gate time properties, we can manage it, then it's up to the user not to shoot themselves in the foot by choosing the wrong one, or by using nonsensical combinations of settings in different instruments.

Another possibility to consider is to move fermatas to the "breaths and pauses" palette (which people ask about periodically anyhow_, and then add a new tenuto marking there that is actually a fermata in disguise (either literally in terms of the internals, or perhaps still an articulation to simplify the layout). Actually, this might be the most sensible way to do it, except it won't help existing scores.

In reply to by Marc Sabatella

YES, YES, YES!!! EXPOSE gate time! That is what I have been wishing for for years. Not only on an individual articulation on an individual note, but default for that articulation type for a score (which is in the data structure already). (Or, better yet, on any note at all).. This would be the greatest improvement since the Piano Editor came to real life, and would make phrasing -de-rigeur- on well-prepared scores.

In reply to by Marc Sabatella

By definition, "Tenuto" isn't the same as a Fermata.

"Tenuto" means to play a note all long the time it should sound (without audible pause with the next).

The other thing is "Rubato", which means some kind of random tempo in some specific note.

Whatever, today MuseScore doesn't play "Tenuto" as well (we can hear a very little pause between notes). It can be seen if you export it as wave file and, then, you see the result in an oscilloscope (or audio editing software, like Audacity).

But... I insist that, in the real world, there are musicians that use the "Tenuto" as a "Legato", at the same time. Even the fact it isn't, academically, right... MuseScore could perform it (because the MIDI system can do it).

In reply to by jotape1960

"Rubato" does not mean "some kind of random tempo in some note". Past participle of "to rob", it means to lengthen one note at the expense of the following notes. See https://en.wikipedia.org/wiki/Tempo_rubato -- I am referring to what it says about Chopin. At any rate, MuseScore should be clear about which markings affect local tempo and which affect note-length relative to notated value and not confuse two,

In reply to by jotape1960

I want to add something about the "Legato".

In the human voice, and some wind instruments, we can "convert" a note in the next (a full linear glissando).

The best sample of this is to hear the very beautiful sound of the electromagnetic instrument called: "Theremin". Please, see: https://www.youtube.com/watch?v=lY7sXKGZl2w

This sound effect is absolutely possible with the MIDI technology and... event the fact it is not always academically right..., MuseScore could play it that way, if it had the full MIDI implementation.

Of course, it is absolutely independent of the visual presentation, which is the main target of the software.

Do you still have an unanswered question? Please log in first to post your question.